Our ultimate goal is to improve our ability to create and measure 3D models derived from cone-beam computed tomography (CBCT). Our main motivation is to improve quality and reduce costs in care of patients with craniomaxillofacial (CMF) deformities. The resulted innovations will also impact other fields. CMF deformities involve congenital and acquired deformities of the jaws and face. A large number of patients in the US and around the world suffered from CMF deformities. The evaluation of these patients includes an assessment of CMF form on 3D models that are traditionally generated from segmented spiral multi-slice CTs (MSCTs). These models are also used to plan their treatment. The purpose of segmentation is to separate different anatomical structures and to remove the artifacts on the CTs. Once 3D models are generated from the segmented CTs, anatomical and teeth landmarks are manually digitized for measurements. Finally, diagnosis and treatment planning are performed based on measurements. Although MSCT provides high- quality images and thus allows relatively fast and easy post processing, many concerns have been raised on excessive radiation exposure to patients. Therefore, more doctors are now using CBCT scanners in their offices. CBCT has less radiation and is inexpensive compared to the MSCT, but their use in generating 3D models is greatly limited by the poor image quality, i.e., low contrast / signal-to-noise ratio and artifacts. Thus, the existing automated segmentation algorithms developed for MSCT are incapable of practically segmenting CBCTs. The current solution to CBCT segmentation entails an arduous and lengthy process that involves labor-intensive manual editing of hundreds of slices. Besides, another arduous and inaccurate task in the assessment of CMF deformities is the digitization of anatomical landmarks on 3D models - the first step to quantify the deformities. Currently a typical 3D cephalometric and teeth analysis requires the manual digitization of more than 200 landmarks, which is time consuming and has limited accuracy. We hypothesize that the creation and measurement of high-quality 3D models can be significantly improved by developing innovative CBCT-friendly post processing tools. Therefore, in this renewal project, we propose to develop and validate a novel CBCT analysis platform to automate the process of CBCT segmentation and landmark digitization. The feasibility of our approaches has already been proven by our preliminary studies. Our innovative CBCT analysis platform will significantly improve the quality and reduce the cost of care to the individuals with CMF conditions. It will change our dental/CMF fields in effectively utilizing CBCT as a guide for on-the-fly diagnosis and treatment planning. With minimal user intervention, the computer will accurately and effectively do the work, which is currently artistically done by the labor-intensive human operators. The resulted innovations may also impact other fields in the future, e.g., orthopedic surgery and cardiovascular surgery where intraoperative whole-body CBCT is acquired for image-guided surgery and intervention.